35
Relationship between body weight and body measurements of Aceh cattle
Putra1*, W.P.B., Sumadi2, Hartatik2, T. and Saumar3, H.
1Laboratory of Animal Breeding and Genetics,Indonesian Institute of Sciences Bogor, West Java 16911, Indonesia. 2Department of Animal Production,Faculty of Animal
Science, Gadjah Mada University, Yogyakarta 55281, Indonesia. 3Indrapuri Breeding and Forages Centre (IBFC) of Aceh Cattle, Aceh 23363, Indonesia.
*Correspondence author: [email protected] Abstract
Data on body weight (BW) and measurements: withers height (WH), body length (BL) and heart girth (HG) of 79 (39 males and 40 females) of Aceh cattle (average 550 days of age) were collected at the Breeding Station in Indrapuri district, Aceh Besar, Indonesia to estimate the BW from body measurements. The data were subjected to standard statistical analysis using SPSS 16.0 software and linear regression analysis was applied by keeping BW as dependent variable and body measurements of WH, BL and HG as the independent variables. The overall means (+
SD) for BW, WH, BL and HG of Aceh cattle over the sexes were 129.37 + 30.84 kg, 96.59 + 7.11 cm, 93.10 + 11.77 cm and 118.15 + 11.08 cm, respectively. Predicted BW based on C, E, F and G regression models showed no significant difference (P > 0.05) with the means of actual BW. The study revealed that HG was the best predictor of BW estimate and this alone contributed more than 80% of the variation in the BW of Aceh cattle. It is concluded that a simple linear regression involving HG as the predictor variable has a high accuracy in predicting BW.
Keywords: Aceh cattle, body measurements, body weight, regression models Introduction
Aceh cattle is one of several Indonesian native cattle breeds that have been the subjects of research and development since 2011 by Indrapuri Breeding and Forages Centre (IBFC) of Aceh Cattle at Indrapuri district, Aceh province, Indonesia. It is a cross between local Zebu cattle, Bos indicus, and Bali cattle since several hundreds years ago (Photograph 1).
This cattle breed is yet to be improved with regard to production performance parameters for higher meat yield under stressful tropical conditions such as low quality nutrient feed, tropical climate, diseases and parasites. As this is the native cattle of Indonesia, there should be intensive genetic improvement by the government (breeding centre) and the
breeders to increase the mean performance of body weight parameters. Most animals are located in the rural areas of Indonesia specifically in Aceh province and owned by rural households, including farmers and minor business owners with small herd size practising traditional feeding and management system.
Often, the marketing of animals is based on visual assessment, while drugs are administrated mostly by estimation, because the use of live weight criterion in feeding, marketing and drug administration requires a weighing scale, which is expensive and not readily affordable by many small rural households. Besides, most farmers do not have the training to use the weighing scales properly. Several studies have been carried out to develop methods of estimating the live
36 body weight of cattle using formulae derived from body measurement (Hardjosubroto, 1994). Body measurements are simple and easily measured variables for estimating the live weight, although the derived prediction equation using body measurements to estimate body weight is unlikely to be more accurate than direct measurement of body weight (using a weighing scale) due to errors in the location of reference points and the anatomical distortions of muscle tone produced when the animal changes position or posture during the procedure of body measurements.
However, body measurements have been used to evaluate breed performance and characterize animals (Warwick et al., 1990),
though general studies have considered only heart girth or may also include body length and withers height in developing predictive equations. The heart girth measurement has been reported to have high correlation with body weight in many breeds of cattle. High correlation between heart girth and body weight were also found in Bali cattle (Gunawan and Jakaria, 2010), Ongole grade cows (Paputungan et al., 2013), male Kamphaengsaen beef cattle (Sawanon et al., 2011) and Tanzania Shorthorn Zebu cattle (Kashoma et al., 2011). This study was therefore aimed at establishing relationship between body measurements and body weight of Aceh cattle for managerial decisions in rural areas of Indonesia.
Photograph 1: A young cow of Aceh cattle Materials and Methods
A dataset of measurements was compiled from 79 Aceh cattle of average age 550 days at the Indrapuri Breeding and Forages Centre (IBFC) of Aceh Cattle, at Indrapuri district, Aceh province, Indonesia.
The cattle sample comprised of 39 males and 40 females. The measurements taken on each animal were body weight (BW) taken using a digital weighing scale, heart girth (HG), measured with a tape measure as
circumference of the chest just behind the foreleg, withers height (WH) measured with a stick-ruler as the distance from the surface of the platform to the dorsal point of the withers and body length (BL) measured using a tape measure from the head of the humerus to the end of the posterior.
Collected data were handled in Microsoft Office Excel 2007 whereas statistical analyses were done using SPSS 16.0 software to generate descriptive statistics. Correlation coefficients between
37 BW and body measurements were obtained.
Linear regression models were fitted with BW as the dependent variable and body measurements (HG, WH and BL) as the independent variables to obtain the relationship between BW and body measurements.
Linear effects of independent variables on body weight were included in the following model (Steel and Torrie, 1980):
Yi = β0 + β1X1 + β2X2 + β3X3 + Ei
where Yi is the body weight (BW) observation of an i-th animal, β0 is the intercept, β1, β2, β3 are the corresponding linear regression coefficients for WH, BL and HG, respectively and Ei is the error term.
The BW was predicted using the prediction equations developed in the present study. Hence, it was posible to make comparisons amongst actual and predicted BW by means of a paired t-test (Sawanon et al., 2011).
Results and Discussion Body Measurements
Aceh cattle were 129.37 kg in body weight at 550 d of age, 96.59 cm in height at
withers measured from the ground and 118.15 cm in height at withers (Table 1).
These were indicators of the size of the animal at 550 days of age. Aceh cattle is one of smallest native breeds in Indonesia. The BW at 550 days of age of several native breeds in Indonesia was reported to be 170.84 + 20.52 kg for Bali, 254.32 + 47.91 kg for Brahman cross, 127.00 + 17.42 kg for Pesisir, 152.30 + 32.00 kg for Ongole crossbreds and 182.02 + 26.60 kg for Katingan cattle (Zurahmah and Teh, 2011;
Duma, 1997; Wirdahayati and Bamualim, 2007; Wijono et al., 2007; Utomo et al., 2010). Abdullah et al. (2007) reported that the BW of Aceh cattle (1.5 years old) were 123.34 + 25.38 kg (male) and 116.70 + 25.83 kg (female). An earlier report by Bakhtiar (2010) revealed the average BW (550 days of age) of Aceh cattle in the same location was 114.75 + 19.66 kg (male) and 113.27 + 16.29 kg (female). The average BW of Aceh cattle in the present study was higher than those reported by Abdullah et al. (2007) and Bakhtiar (2010). These differences might be due to the variation in genetic effects, environmental differences in management practices from year to year (Falconer and Mackay, 1996).
38
Table 1. Mean body weight and body measurements of Aceh cattle
Variables Mean SD Range CV (%)
Male (n=39)
BW (kg) 131.23 27.28 67.00 – 183.00 20.79 WH (cm) 98.41 7.92 84.00 – 132.00 8.05 BL (cm) 94.72 14.56 78.00 – 172.00 15.37 HG (cm) 117.38 9.79 91.00 – 132.00 8.34 Female (n=40)
BW (kg) 127.55 34.21 71.00 – 223.00 26.82 WH (cm) 94.83 5.78 82.00 – 105.00 6.10 BL (cm) 91.53 8.09 73.00 – 103.00 8.84 HG (cm) 118.90 12.29 96.00 – 145.00 10.34 Total (n=79)
BW (kg) 129.37 30.84 67.00 – 223.00 23.84 WH (cm) 96.59 7.11 82.00 – 132.00 7.36 BL (cm) 93.10 11.77 73.00 – 172.00 12.64 HG (cm) 118.15 11.08 91.00 – 145.00 9.38 LW: body weight; WH: withers height; BL: body length; HG: heart girth;
SD: standard deviation; CV: coefficient of variation; n = number of observations
Correlation and Coefficients of Determination
Positive correlation was found between body measurement parameters and BW in Aceh cattle indicating as the body measurements increase the body weight also increases. Among these three measurements, HG had the highest correlation coefficient with BW (0.93 for male and 0.84 for female).
The correlation coefficients between HG and BW in several breeds of cattle such as Tanzania Shorthorn Zebu (0.94), Kamphaengsaen (0.91), Nyalawi (0.88),
Nguni (0.58), Holstein (0.78), Brown Swiss (0.98), Bali (0.87) and Ongole crossbred (0.48) have been reported (Kashoma et al., 2011; Sawanon et al., 2011; Alsiddiq et al., 2010; Serkan and Yalzin, 2009; Gunawan and Jakaria, 2010; Wijono et al., 2007).
Nesamvuni et al. (2000) reported that the correlation coefficient between HG and BW in male Nguni cattle (0.76) was higher than female cattle (0.62) similar to the finding of the present study. The high correlation between BW and body measurements suggest that HG could provide a good estimate of BW of Aceh cattle.
39
Table 2. Pearson correlation coefficients between variables of Aceh cattle
Variables BL HG BW
Male (n=39)
WH 0.87* 0.38 0.48*
BL - 0.12 0.26
HG - - 0.93*
Female (n=40)
WH 0.82* 0.80* 0.73*
BL - 0.84* 0.81*
HG - - 0.84*
n = number of observation; * = significant value at P < 0.05
Table 3 presents a summary of the simple and multiple linear regression analyses and the models generated from predicting BW from the linear body measurements. Based on the regression models (A, B and C) BW could be predicted using parameters that had high coefficient of determination (R2). The equation for BW from body measurement of HG alone (C models) was Y = -172.47 + 2.59 HG (R2=0.86) for males and Y = -148.68 + 2.32 HG (R2=0.70) for females. This showed that when increasing HG by 70 cm in males and females, the corresponding increase in BW could be about 8.83 kg and 13.72 kg, respectively. Kashoma et al. (2011) reported that the coefficient of determination (R2)
based on the C models in Tanzania Shorthorn Zebu cattle were 0.88 (males) and 0.87 (females), also Milla et al. (2012) reported in Nilotic cattle were 0.95 (males) and 0.94 (females).
The parameter estimates in the multiple linear regression models showed that more than one body measurements may be required to predict the BW in Aceh cattle.
In the present study WH, BL and HG were the important body measurements required for predicting the BW of Aceh cattle, based on the highest R2 values of the multiple linear regression equations. However, the accuracy of estimation could be improved if the variables were combined in a multiple regression.
40
Table 3. Simple and multiple regression models for predicting body weight (dependent variable) from linear body measurements (independent variable) of Aceh cattle
Models Independent
variables Intercept
Regression coefficient
SE r R2
WH BL HG
Male (n=39)
A WH -31.27 1.65 - - 24.26 0.48 0.23
B BL 85.74 - 0.48 - 26.72 0.26 0.07
C HG -172.47 - - 2.59 10.29 0.93 0.86
D WH; BL -110.26 3.66 -1.26 - 22.81 0.58 0.34
E WH; HG -204.98 0.52 - 2.43 9.67 0.94 0.88
F BL; HG -192.59 - 0.27 2.54 9.61 0.94 0.88
G WH; BL; HG -197.89 0.18 0.19 2.50 9.73 0.94 0.88 Female (n=40)
A WH -279.73 4.30 - - 23.83 0.73 0.53
B BL -186.13 - 3.43 - 20.32 0.81 0.66
C HG -148.68 - - 2.32 19.08 0.84 0.70
D WH; BL -233.77 1.15 2.76 - 20.21 0.82 0.67
E WH; HG -199.37 1.01 - 1.95 19.00 0.84 0.71
F BL; HG -189.25 - 1.57 1.46 18.00 0.86 0.74
G WH; BL; HG -195.42 0.15 1.51 1.43 18.24 0.86 0.74 WH: withers height; BL: body length; HG: heart girth; r: correlation coefficient;
R2: coefficient of determination; SE: standard error of the estimate; n = number of observation Table 4 shows that the C, E, F and
G models produced no significant (P > 0.05) difference between actual and predicted body weight for each group. Although when compared with the E, F and G models, the C models had lower R2 value for each group (Table 3), the input parameters required can
be measured using only HG, which is easy and fast for the farmer, especially for male Aceh cattle (R2= 0.86). Similar findings reported by Sawanon et al. (2011) for C models had high R2 value (0.83) in male Kamphaengsaen cattle.
41
Table 4. Difference between the actual body weight and the predicted body weight using several regression models
Models Mean (kg) SD Range CV (%) Sig.*
Male (n=39)
A 131.26 13.01 107.33 – 186.53 9.91 S
B 131.21 6.98 123.18 – 168.30 5.32 S
C 131.59 25.31 63.22 – 169.41 19.24 NS
D 130.89 15.67 97.97 – 162.56 11.97 S
E 131.52 25.57 60.87 – 170.55 19.44 NS
F 131.18 25.60 59.88 – 169.96 19.52 NS
G 131.33 25.61 60.10 – 169.84 19.50 NS
Actual 131.23 27.28 67.00 – 183.00 20.79 -
Female (n=40)
A 128.39 24.76 72.97 – 171.77 19.29 NS
B 127.84 27.72 64.26 – 167.16 21.68 NS
C 127.26 28.58 74.04 – 187.72 22.46 NS
D 128.02 28.02 63.16 – 165.74 21.88 NS
E 128.42 28.88 75.70 – 189.43 22.49 NS
F 128.11 29.46 69.90 – 181.02 22.99 NS
G 127.12 29.37 68.83 – 180.19 23.11 NS
Actual 127.55 34.21 71.00 – 223.00 26.82 -
SD: standard deviation; CV: coefficient of variation; S: significant at P < 0.05;
NS: not significant at P > 0.05 Conclusion
The strong relationship between body weight and linear body measurements of Aceh cattle indicated that these variables or their combination could be used to estimate the body weight of these cattle.
Heart girth had the highest correlation with body weight for each group. Simple and multiple linear regression models using other body measurement parameters that had high coefficient of determination (R2 >0.80) could also be utilized. The C, E, F and G models for male cattle had the high R2 value and these predicted equations could be used to predict the body weight of Aceh cattle.
Acknowledgement
The authors expressed appreciation to the Indrapuri’s Breeding and Forages
Centre (IBFC) of Aceh Cattle at Indrapuri district, Aceh Besar Regency, Aceh province, Indonesia and to the barn employees who were involved in the collection of data.
References
Abdullah, M.A.N., Noor, R.R., Martojo, H., Solihin, D.D. and Handiwirawan, E.
2007. The phenotypic variability of Aceh cattle in Nanggroe Aceh Darussalam. J. Indon. Trop. Agric. 33:
11-21.
Alsiddiq, M.A., Babiker, S.A., Galal, M.Y.
and Mohammed, A.M. 2010.
Phenotypic characterization of Sudan Zebu cattle (Baggara type). J. Anim.
Vet. Sci. 5: 10-17.
42 Bakhtiar. 2010. The performance of
production and reproduction in Aceh Cattle. MSc. Thesis. Andalas University, West Sumatera, Indonesia.
Duma, Y. 1997. Estimation of genetic parameters in Brahman cross and Ongole cattle at Bila River Ranch pasture. MSc. Thesis. Gadjah Mada University, Yogyakarta, Indonesia.
Falconer, D.S. and Mackay, T.F.C. 1996.
Introduction to Quantitative Genetics.
4th Edition. Longman Group, Harlow, England.
Goe, M.R., Alldredge, J.R. and Light, D.
2001. Use of heart girth to predict body weight of working oxen in the Ethiopian highlands. Livest. Prod. Sci. 69: 187- 195.
Gunawan, A. and Jakaria. 2011. Application of linear body measurements for predicting weaning and yearling weights in Bali cattle. J. Anim. Prod. 12: 163- 168.
Hardjosubroto, W. 1994. Application of Animal Breeding. P.T. Gramedia Widiasarana, Jakarta, Indonesia.
Kasoma, I.P.B., Luziga, C., Werema, C.W., Shirima, G.A. and Ndossi, D. 2011.
Predicting body weight of Tanzania Shorthorn Zebu cattle using heart girth measurements. Livesttock Research
for Rural Development.
http://Irrd.org/Irrd23/4/kash23094.htm.
Milla, A.P., Mahagoup, M.M.M. and Bushara, I. 2012. Estimation of live body weight from heart girth, body length and in condition score in Nilotic cattle – Southern Sudan. J. Anim. Sci.
Adv. 2: 253-457.
Nesamvuni, A.E., Mulaudzi, J., Ramanyimi, N.D. and Taylor, G.J. 2000. Estimation of body weight in Nguni-type cattle under communal management condition. S. Afric. J. Anim. Sci. 30: 97- 98.
Paputungan, U., Hakim, L., Ciptadi, G. and Lapian, H.F.N. 2013. The estimation accuracy of live weight from metric body measurements in Ongole grade cows. J. Indon. Trop. Anim. Agric. 38:
149-155.
Sawanon, S., Phoompong, B. and Preecha, I.
2011. Body measurements of male Kamphaengsaen beef cattle as parameters for estimation of live weight.
Kasetsart J. Nat. Sci. 45: 428-434.
Serkan, O. and Yalcin, B. 2009. The accuracy of prediction of body weight from body measurements in beef cattle.
Arch Tierz. 52: 371-377.
Steel, R.G.D. and Torrie, J.H. 1980. Principle and Procedure of Statistics: A Biometrical Approach. 2nd Edition. Mc- Grawhill International Book Inc., Toronto, Canada..
Utomo, B.N., Noor, R.R., Sumantri, C., Supriatna, I. and Gunardi, E.D. 2010.
Morphometric performances of Katingan cattle in central Kalimantan. J.
Indon. Anim. Vet. 15: 220-230
Warwick, E.J., Astuti, J.M. and Hardjosubroto, W. 1990. Animal Breeding. Gadjah Mada University Press, Yogyakarta, Indonesia.
Wijono, D.B., Hartati and Dicky, M.D. 2007.
Correlation of body size with growth rate of live weight of Ongole crossbred cattle. Proceedings of the National Animal and Veterinary Technology Conference, Bogor, Indonesia pp. 236- 239.
Wirdahayati, R.B. and Bamualim, A. 2007.
The productivity of Pesisir cattle and the carrying capacity of communal grazing land in the district of Pesisir Selatan in West Sumatera. Proceedings of the National Seminar of Animal Science and Veterinary Technology. Bogor, Indonesia pp. 122-131.
43 Zurahmah, N. and Teh, E. 2011. The
estimation of body weight of bulls candidate of Bali cattle using some body dimension. Bullet. Anim. Sci. 35: 160- 164.
44
